RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen3.5 35B A3B run on the 2x RTX 3090?

Yes. At 8K context the best fit is Q8_0, using about 35.4 GB of the 45.6 GB usable on the 2x RTX 3090. Estimated speed ceiling: ~304 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K)13.5 GB Fits13.6 GB Fits14.1 GB Fits15.9 GB Fits832
3-bit (Q3_K_M) •16.2 GB Fits16.3 GB Fits16.7 GB Fits18.6 GB Fits686
4-bit (Q4_K_M) •21.5 GB Fits21.5 GB Fits22.0 GB Fits23.9 GB Fits509
5-bit (Q5_K_M) •25.4 GB Fits25.5 GB Fits25.9 GB Fits27.8 GB Fits427
6-bit (Q6_K) •27.8 GB Fits27.9 GB Fits28.4 GB Fits30.2 GB Fits389
8-bit (Q8_0) •35.3 GB Fits35.4 GB Fits35.9 GB Fits37.7 GB Fits304
16-bit (F16/BF16) •65.6 GB No65.6 GB No66.1 GB No68.0 GB No—

• real GGUF file size from unsloth/Qwen3.5-35B-A3B-GGUF. Speed is a bandwidth ceiling (936 GB/s), not a benchmark. Methodology.

More

Qwen3.5 35B A3B on every GPU and Mac · Every model that runs on the 2x RTX 3090 · Try other settings in the calculator

Data: Hugging Face Hub (Qwen/Qwen3.5-35B-A3B), NVIDIA specs. Data as of 2026-11-17.